Why Consumption Doesn't Swing When Income Shocks Hit
The capacity to smooth out shocks to income and avoid large swings in consumption pattern is an important determinant of individual households' welfare. Nicola Pavoni (Department of Economics) and Orazio Attanasio (University College London) in their recent Risk Sharing in Private Information Models with Asset Accumulation: Explaining the Excess Smoothness of Consumption (Econometrica, Vol. 79, No. 4, July 2011, 1027-1068, doi: 10.3982/ECTA7063) study both the cross-sectional properties and the inter-temporal dynamics of consumption allocation. The authors model the households' decision over their consumption path and show that the predictions of the model are consistent with what is observed in the data, in that consumption patterns react less than one-to-one to permanent innovations to income and do not react unduly to predictable changes in income.
More specifically, the study focuses on an economic environment in which agents have unobservable private information on the two elements that constitute income; namely, the effort they put in work and shocks to productivity. Furthermore, consumption of individuals is presumed to be unobservable, which renders their savings unobservable as well. As such, agents can allocate their savings to a risk free asset and/or insure themselves against idiosyncratic income shocks by entering asset-trading arrangements across states of the world that allow them to obtain state-contingent transfers. The authors show that in such a set up consumption allocations exhibit the so-called "excess smoothness" hypothesis in the sense that individuals obtain more insurance against income shocks in comparison to a self-insurance economy in which agents have access to a single asset of fixed return only.
Authors make a challenging attempt to bring together two strands of the literature and analyze the elements of consumption smoothing as well as insurance arrangements against idiosyncratic shocks in asset markets that are incomplete due to the presence of moral hazard. Therefore, within a unified framework, the links between private savings and insurance under moral hazard are brought together. In addition, the authors highlight how some of the empirical results in the consumption literature could be understood, through the use of their model, as evidence of partial insurance and on the pertinence of specific imperfections in the asset market.
The first contribution of the paper lies in a theoretical ground. In a model with moral hazard and hidden saving, excess smoothness of consumption is observed due to the accessibility of additional insurance in comparison to an economy with a single asset that returns a fixed risk-free rate. Moreover, the authors construct specific examples from which closed-form solutions can be derived, which, in turn, help relate the excess smoothness coefficients to the structural parameters of the model.
The second contribution of the study is of empirical nature. The authors take their model to data and provide a test of excess smoothness of consumption on a pseudo-panel from the United Kingdom. They first focus on risk sharing across cohorts and complement this approach with a test based on movements in the cross-sectional variance of consumption, which stresses insurance within cohorts. It turns out the authors' empirical findings are remarkably in line with the predictions of their model.